DIY Video Projector

Status
Not open for further replies.
Jump to Latest
#1,784
projector

well
I have to appologise to everyone. I went and bought a projector off of ebay. But the plus side is it is a fully operational proxima dp 2400 with two good bulbs and it only cost me $240us with shipping.

I still plan on building a pj and this will give me a reference point to shoot for in quality and brightness.

Besides, I could easily spend more than $250us putting one together from spare bits and pieces.
 
#1,785
Hey, I've got my projector up and running. It works great (though it is just the OHP method). The only problem is bulb life. The bulbs my overhead uses only last 35 hours and at around $20 per bulb thats pretty expensive. Do you guys know if a longer lasting bulb that will work with my overhead exists? The bulb type im using is "ELH". It is 300w 120v.

Here's what my bulb looks like:

elhpic6.jpg
 
#1,786
As far as the LED into fiber optic is concerned: maybe this is stupid, but couldn't you make the light from the LEDs more parallel by flattening the round tops of the plastic casings? For an average LED, this would allow about a 60 degree cone of light to reach the flat top, and by Snell's law (with plastic having a rough n value of 1.5 and air 1.0) the result would be about a 97 degree cone of light leaving the LED. If the distance between the the tip of the casing and the diode was increased, the spread of light could be reduced. Or, by replacing the air/plastic interface with a water/plastic interface, the light cone could be reduced to about a 70 degree spread. The useable light from the diode would be a lot less, and the light coming from the sides of the casings would still be an issue... Just a thought I guess.
 
#1,787
another problem

I had another problem (so many problems, that's why this thread come alive), when I used DCX or PCX lens projecting image (7"x9" source image), I couldn't get the whole image sharp on the screen. Every time I got sharp in the middle all the 4 side became blurr, viceversa: sharp at 4 side, blured in the middle. Why?
Anybody had the same problem as me? or somebody know the answer?
 
#1,788
LEDs will work

Hi everyone,

I have been lurking here for 3 days having picked up the trail from /. and trying to catch up with all of you prolific posters! Have learned a lot from all of you. Thank you for all of the excellent links on where to shop for cheap optics. SuperDave: thank you for demonstrating that you can use a 15" LCD and get those pesky controller boards out of the way. This is the route I wanted to take since it is the only way that will give XGA resolution and good CR. Thank you Vince for starting this truely amazing thread that has taken me several days to read!

Right now I am going for the ultra-cheap approach to see if I can get anything to work. I am using a 225W Halogen (6500 lumens) in a 500W work lamp that I bought from Office Depot. The condenser lense is two Fresnel lenses (really "page enlargers" from Staples). My LCD is just a transparency from a talk I gave last year. I plan to pick up a lens from suplusshack but haven't decided what to use yet. I will probably start with something simple and switch to an acromat later. Total cost of the project so far: $50 US. My beam path looks like this:

halogen -> FL1 -> FL2 -> "LCD" -> final lens

The reason I am writing now is that I just wanted to say that LEDs will work. You just have to use a lot of them and make an array that is the same size as the LCD. The brightest white LED source is from Luxeon (just announced in a press release but you cannot buy it yet). It is a 5-Watt 120 lumen source. The package is 20mm x 20mm (in the shape of a hexagon). So just make a hexagonal close packed array of these bad boys. You can fit 200 of them for a total brightness of 24K lumens. Of course, it will require 1000W of power and may turn out to be prohibitively expensive.
😉

Good luck to all

CR
 
#1,789
reflecting light

Fender, remp....What do you mean I cant focus both reflected and non-reflected light. I am not a light/lens expert by any means and this is where I need the most information so I am asking based on my current understanding and would like to know if I'm wrong.

I did not know there was any issue with focusing both reflected and non-reflected light. I knew that there were issues of non-polarized light going through a polarizer but since I am not even using a polarizer (I am not using my LCD right now) I though the fresens would be able to pick up and focus ALL light that goes towards it.

I have used my fresnels to focus a small (real small cuz I was losing a lot to scatter) amount of light before I was even using a reflector. And...when I started using one I got a lot more light focuses....So that leads me to believe that I am able to focus both reflected and non-reflected light. If I am way off here (and you are sure of yourself) please tell me. Also I (and I am sure others) would like to know some good links explaining light and optics properties...all I have found is some pretty dry textboob stuff...oh well).

Anyways... I thought that when using fresnels that the fisrt fresnel would pick up ALL refelcted and non-reflected light from ANY angle and send it out in a parallel beam. The second fresnel was there to condense the light down to the projection lens. I thought that the reflector was just used to add more light (so light from all directions from the lamp eventually hit the fresels).

Please explain/help....Thank,

Dave
 
#1,790
focusing

SuperDave

A lens bends any light that hits it to a new angle.
consider the following:
parabolic reflecter with a point light source at the focal point.
light comming directly from the light will be diverging, light from the reflector will be parallel.
when both "types" of light hit the lens the parallel light will now be converging and the direct light will be parallel( assuming that the light source is at the focal lenght of the lens.)

hope that helps.
 
#1,792
Now, if you could get one of those LCDs, we would have a lot less problems with light. Remember .... Very old LCD projectors use 6.5" or hogher LCD panels and FXL lamps (smae used on 4000 lumens OHP) with resolution of 640x480 and contrast of 100:1.

What we are accomplishing now with commercial LCD panles is more quality than those earlier projectors. Hey, we can even pair with the quality even just using a OHP with a panel on it.

For those who think we can't build a quality one, Id think twice. We sure do have problems with light but manufactors also have problems with that too. Just look how low lumens those earlier projectors have (250-300 from a 4000 lumens lamp)


I think that even if we build a perfect reflector and use wave retarders with polarizer, we will still not be able to improve much. We need to reduce the size of the LCD panels while still keep the high resolution we are using now (at least 800x600). This is the path manfucators used to improve their projectors and this is how we got to do.

This way we could produce projectors with similar quality at lower cost to build and a lot less money to maintain as bulb replacement could be a lot less and less parts.

I doubt we will ever be able to perfect a reflector and make huge improvement. Im pretty sure manufactors tried it before. The way to go is finding a small high-resolution lcd. Trying different lamps and working around the reflector won't solve the problem.
 
#1,793
BRUTE FORCE

I have decided to brute force my way through the rest of this project. I'm going to try my best to make a reflector out of the aluminum that I have polished. I'm sure it will help quite a bit. However, I'm going to put a 400W Metal Halide Bulb in front of it. Ahh yes. well over 35k lumens. Whatever I can't reflect will just not be used, and its as simple as that. As many have already said, theres not much we can do. We need very large high quality optics, precision reflectors, and bulbs with extraordinary brightness with the tiniest of arcs.

I'm going to order a ballast for about $60 shipped from the place you guys have mentioned. I called up the guy, he was really cool about it, and seems nice.

http://lighting.webpipe.net/ballast.htm

I still need a mogul base and bulb. For bulb, I have been thinking

http://www.goodoffices.com/store/product.asp?productid=36795

39,000 lumen, 20k hours, 4200K color temp. couple reasons I chose this were its size (its about 3 inches smaller than the ED37 Mercury vapor bulbs we were playing with), and its output, which seemed pretty high for a 400W. who knows, I might just end up picking up a 400W at Home Depot tonight. I'm going to see about the plas-tex stuff to make a screen, and buy a drill since I need one.
 
#1,794
For those of you going with the OHP method, have you been able to find a bulb with a decent life time? The bulb im using for my overhead only lasts 35 hours (its an ELH, 120v 300w). Are there any higher quality bulbs that last maybe 200 or more hours? a 1000 or 10000 hours would be nice too! Thanks.
 
#1,795
bulb improvement for OHP

BSM;

You might check out the posts a few back with the pix of the Dukane Quantum - it looks to be the same optical path as a 4003, only with a MH lamp (and a ballast) wedged in there. That MH lamp might drop in in place of the incandescent your projector uses. This would increase the time to failure by better than 10X -

You could get a ballast to light it from the same place as Undream -
 
#1,796
Undream

That MH bulb you are looking at has an arc lenght of 1.54 inches too long to be regarded as a point source so you will have trouble using a normal reflector. There is another reflector you could use which is designed for long arcs specially horizontal laying bulbs which is called an orthogonal parabolic reflector. It collects all the 90 degree rays from the bulb and focuses them to a point much like an elliptical. The rays carry on past the point so they could illuminate your fresnel. This reflector was used by Japanese projector manufacturers until they moved over to short arc lamps. You can see a detailed description and drawing at US patent database patent #6322219 figure 16. You could use my laser reflector shaper gizmo thing posted page or two back to help get the reflector shape.
Hope this helps.
 
#1,798
Undream


What about the bulb from the $100 Metal Halide project page?

http://www.hellolights.com/400wat65iw.html

It's very bright; I'm working on an ellipsoidal reflector for it.


ZeCo

The $80 camera in this link has a QVGA LCOS which can be used standalone or can be multiplexed for a higher resolution display.


http://shop3.outpost.com/product/3131320

Further info on the display is available here. They will sell you a kit for $2000+, so the camera is the way to go.

http://www.displaytech.com/products/personal_view/mv2.html
 
#1,799
YES!

Orthogonal parabolic reflector! Thanks, remp...I've been thinking of something like that since the good 'ol days of messing with fluorescents, but I didn't know there was a name for it.

That would be a BIG improvement over elliptical/parabolic designs that we are looking for. No matter how carefully you design one of those reflectors, there will still be only one true focal POINT...with standard MH of MV bulbs, most of the arc will be relatively far from that focal point. Only the light from the reflector's true focal point will be reflected parallel (or converging to the second focal point in the case of an elliptical reflector).

Sorry if I am stating the obvious, but I feel like it really needs to be addressed. Thanks for pointing it out, remp!

Now, how do I calculate an orthogonal elliptical reflector......😕

-f4
 
Status
Not open for further replies.
Top Bottom